Densely Connected Convolutional Networks for Speech Recognition

2018/08/10 by Li, Chia Yu, Vu, Ngoc Thang
#Computation and Language (cs.CL) #FOS: Computer and information sciences

paper · doi:10.48550/arxiv.1808.03570

Abstract

This paper presents our latest investigation on Densely Connected Convolutional Networks (DenseNets) for acoustic modelling (AM) in automatic speech recognition. DenseN-ets are very deep, compact convolutional neural networks, which have demonstrated incredible improvements over the state-of-the-art results on several data sets in computer vision. Our experimental results show that DenseNet can be used for AM significantly outperforming other neural-based models such as DNNs, CNNs, VGGs. Furthermore, results on Wall Street Journal revealed that with only a half of the training data DenseNet was able to outperform other models trained with the full data set by a large margin.

Related

• Investigation of Densely Connected Convolutional Networks with Domain\n Adversarial Learning for Noise Robust Speech Recognition
• Fully Convolutional Speech Recognition
• Deep Neural Networks for Acoustic Modeling in Speech Recognition: The Shared Views of Four Research Groups
• QuartzNet: Deep Automatic Speech Recognition with 1D Time-Channel Separable Convolutions
• Very Deep Convolutional Neural Networks for Robust Speech Recognition